复杂振动环境下磁悬浮陀螺全站仪数据处理算法研究

发布时间：2018-06-08 17:23

本文选题：磁悬浮陀螺 + 转子电流　；参考：《长安大学》2017年硕士论文

【摘要】：高精度磁悬浮陀螺全站仪利用磁悬浮技术作为陀螺灵敏部的支承,突破了传统悬挂带式支承技术的瓶颈,并可以无依托、自动快速的测定真北方位,确定任意方向线的坐标方位角。由于其具有自动化程度高,观测时间短,定向精度高,环境适应性强等优点,而被广泛应用于现代大型地下工程测量和科学研究中。磁悬浮陀螺全站仪属高敏感性仪器,环境的不稳定性对陀螺定向的影响不容忽视,如何克服环境中振动等复杂因素的影响是本文研究的重点。论文从磁悬浮陀螺数据处理上来提高定向精度,主要研究内容包括磁悬浮陀螺转子电流数据滤波和时频幅值谱分析。首先分析转子电流数据的分布特性,并对在不同振动环境中采集的转子电流数据进行分类,然后利用希尔伯特-黄变换(HHT)、无迹卡尔曼滤波(UKF)和粒子滤波(PF)方法对分类出的每一类数据进行处理,这三种方法均是处理非线性非平稳信号较有效的算法。HHT是基于EMD分解出有限阶的IMF函数,并对每一阶IMF建立能量分布图,判断处噪声与信号的主导模态,对转子电流信号进行重构;UKF是对转子电流数据进行UT变换后,再利用卡尔曼滤波处理;粒子滤波处理时选取不同的粒子数,对比不同粒子数下的处理结果。将三种滤波的处理结果进行对比分析,总结出不同波形分布的转子电流数据对应的最优算法。基于HHT方法的Hilbert时频幅值谱和边际谱来分析不同波形分布的转子电流数据,通过边际谱中高幅值对应的频率来分析信号中频率成分的存在,判断时频幅值谱中是否存在贯穿于整个采样过程的明显边频带,最终来探索转子电流数据波形分布与采样时环境振动的关系。
[Abstract]:The high precision magnetic levitation gyroscope total station uses magnetic levitation technology as the support of the sensitive part of the gyroscope. It breaks through the bottleneck of the traditional suspension belt support technology, and can automatically and quickly determine the true north azimuth and determine the coordinate azimuth of any direction line without relying on it. Because of its high degree of self movement, short observation time, high orientation precision and environment With the advantages of adaptability and so on, it is widely used in modern large-scale underground engineering measurement and scientific research. Magnetic levitation gyroscope total station is a Gao Min perceptual instrument. The influence of environmental instability on Gyroscopic Orientation can not be ignored. How to overcome the influence of complex factors such as vibration in the environment is the focus of this paper. The number of magnetic levitation gyroscopes is from the number of magnetic levitation gyroscopes. The main research contents include the rotor current data filtering and time frequency amplitude spectrum analysis of magnetic suspended gyroscope. Firstly, the distribution characteristics of rotor current data are analyzed, and the rotor current data collected in different vibration environments are classified, and then Hilbert yellow transform (HHT) and Untraced Calman filter are used. (UKF) and particle filter (PF) method to deal with each class of data. These three methods are all the effective algorithms to deal with nonlinear non-stationary signals..HHT is based on the EMD decomposition of the finite order IMF function, and the energy distribution map of each order IMF is set up to judge the dominant mode of noise and signal, and the rotor current signal is weighed. UKF is the UT transformation of the rotor current data, and then the Calman filter is used, and the number of particles in the particle filter is selected to compare the processing results under different number of particles. The results of the three filtering are compared and analyzed, and the optimal algorithm for the corresponding rotor current data with different waveform distribution is summed up. Based on the HHT square The Hilbert frequency amplitude spectrum and the marginal spectrum of the method are used to analyze the rotor current data of different waveforms, and the existence of the intermediate frequency components of the signal is analyzed by the corresponding frequency of the high amplitude in the marginal spectrum, and whether there is a clear side frequency band that runs through the whole sampling process in the time frequency amplitude spectrum, and finally to explore the distribution of the rotor current data waveform. The relationship between the vibration of the environment and the sampling.
【学位授予单位】：长安大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TB22

【参考文献】